John, (01)
On Mon, Feb 1, 2010 at 7:54 PM, John F. Sowa <sowa@xxxxxxxxxxx> wrote:
> ...
> There is no such thing as a one-size-fits-all ontology that
> can cover chess, hiking, programming, eating, cooking, chemistry,
> surgery, driving a car, and football. For most of those subjects,
> a very large part of the background knowledge will *not* be
> expressed in either verbal patterns or some version of logic. (02)
No contest on the impossibility of a "one-size-fits-all ontology". (03)
Also no contest on the need for "background knowledge". (04)
But why can't "background knowledge" too be represented in patterns
among text, patterns of word use etc? (05)
I agree the "background knowledge" must be found outside any simple
sentence being interpreted. That goes without saying. You would have
to find patterns in the whole language. (06)
Think of it as a machine learning problem. I'm talking about learning
a whole ontology, theory, even logic, directly from raw patterns in
text, then using it to interpret a sentence in context. (07)
The machine learning problem has been attempted before. The new thing
is that we agree anything you learn must be partial. Extremely partial
I believe. Specific to each sentence, or more. That is what makes this
machine learning effort different from all earlier efforts. But, given
that we've agreed whatever you learn must be partial, what will be the
limits on what can be learned? (08)
-Rob (09)
P.S. Ron. From a cursory reading your interest seems to be in the
grounding of ontology more directly in human perception. This is an
approach I sympathize with, for my usual reason, that abstractions
have limits. I don't personally think it is necessary, but I
sympathize. Anyway, it reminded me of a maths reference you might
like. A French mathematician who believes maths is inseparable from
human biology: (010)
Mathematical Intelligence, Infinity and Machines: Beyond Gödelitis
Longo G. CNRS and Dépt. De Mathématiques et Informatique, Ecole
Normale Supérieure, Paris, France
We informally discuss some recent results on the incompleteness of
formal systems. These theorems, which are of great importance to
contemporary mathematical epistemology, are proved using a variety of
conceptual tools provably stronger than those of finitary
axiomatisations. Those tools require no mathematical ontology, but
rather constitute particularly concrete human constructions and acts
of comprehending infinity and space rooted in different forms of
knowledge. We shall also discuss, albeit very briefly, the
mathematical intelligence both of God and of computers. We hope in
this manner to help the reader overcome formalist reductionism, while
avoiding naïve Platonist ontologies, typical symptoms of Gödelitis
which affected many in the last seventy years. (011)
ftp://ftp.di.ens.fr/pub/users/longo/PhilosophyAndCognition/incompl-inf.dvi.pdf (012)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (013)
|